| Structural health monitoring systems can identify structural conditions and alarm structural damage in time,which can avoid accidents,casualties,and property losses.The electromechanical impedance(EMI)method is a kind of low-cost and high-sensitivity structural health monitoring method.As for damage identification,conventional data-driven based methods in EMI methods relate the structural condition to the change between baseline and measured responses.It is difficult to obtain suitable baselines in practical work environment.Aiming at above problems,this paper proposed an EMI-based damage identification method that uses tunable inductive shunt circuity for data enrichment.The main contents are as follows:(1)Aiming at the measurement structure based on a single piezoelectric transducer,a novel method is proposed by using dual-piezoelectric transducers.The equations of motion and simulation of the designed measurement structure in the finite element form are deduced and performed,which proves the effectiveness of the proposed method in decoupling the impedance of the structure from EMI.(2)A data enrichment method that uses tunable inductive circuity is proposed for EMI-based damage identification systems.A SHM system based on the phase-sensitive detection algorithm is designed to acquire system frequency responses.The tunable inductive circuity can effectively enrich the training dataset for damage detection by introducing a new adjustable degree of freedom to the system while keeping the same mechanical structure.This method is a potential approach to be applied in data-driven based structural health monitoring system.(3)One dimensional convolutional neural network and deep transfer learning model are introduced to classify the composite faults of bolt looseness and mass loss.Experiments verify that the utilized transfer model is effective for structural health monitoring systems.The accuracy of the collected dataset achieves 99.24%. |
PDF Full Text Request